84d9ce30cb
Add option for float32 sampling with float16 UNet
...
This also handles type casting so that ROCm and MPS torch devices work correctly without --no-half. One cast is required for deepbooru in deepbooru_model.py, some explicit casting is required for img2img and inpainting. depth_model can't be converted to float16 or it won't work correctly on some systems (it's known to have issues on MPS) so in sd_models.py model.depth_model is removed for model.half().
3 years ago
c1928cdd61
bring back short hashes to sd checkpoint selection
3 years ago
a5bbcd2153
fix bug with "Ignore selected VAE for..." option completely disabling VAE election
...
rework VAE resolving code to be more simple
3 years ago
08c6f009a5
load hashes from cache for checkpoints that have them
...
add checkpoint hash to footer
3 years ago
febd2b722e
update key to use with checkpoints' sha256 in cache
3 years ago
f9ac3352cb
change hypernets to use sha256 hashes
3 years ago
a95f135308
change hash to sha256
3 years ago
4bd490727e
fix for an error caused by skipping initialization, for realsies this time: TypeError: expected str, bytes or os.PathLike object, not NoneType
3 years ago
1a23dc32ac
possible fix for fallback for fast model creation from config, attempt 2
3 years ago
4fdacd31e4
possible fix for fallback for fast model creation from config
3 years ago
0f8603a559
add support for transformers==4.25.1
...
add fallback for when quick model creation fails
3 years ago
ce3f639ec8
add more stuff to ignore when creating model from config
...
prevent .vae.safetensors files from being listed as stable diffusion models
3 years ago
0c3feb202c
disable torch weight initialization and CLIP downloading/reading checkpoint to speedup creating sd model from config
3 years ago
552d7b90bf
allow model load if previous model failed
3 years ago
642142556d
use commandline-supplied cuda device name instead of cuda:0 for safetensors PR that doesn't fix anything
3 years ago
68fbf4558f
Merge remote-tracking branch 'Narsil/fix_safetensors_load_speed'
3 years ago
0cd6399b8b
fix broken inpainting model
3 years ago
8d8a05a3bb
find configs for models at runtime rather than when starting
3 years ago
02d7abf514
helpful error message when trying to load 2.0 without config
...
failing to load model weights from settings won't break generation for currently loaded model anymore
3 years ago
8f96f92899
call script callbacks for reloaded model after loading embeddings
3 years ago
311354c0bb
fix the issue with training on SD2.0
3 years ago
f55ac33d44
validate textual inversion embeddings
3 years ago
5ba04f9ec0
Attempting to solve slow loads for `safetensors`.
...
Fixes #5893
3 years ago
3bf5591efe
fix F541 f-string without any placeholders
3 years ago
5a650055de
Removed lenght in sd_model at line 115
...
Commit eba60a4https://github.com/AUTOMATIC1111/stable-diffusion-webui/issues/5971#issuecomment-1364507379
3 years ago
eba60a42eb
Merge pull request #5627 from deanpress/patch-1
...
fix: fallback model_checkpoint if it's empty
3 years ago
ec0a48826f
unconditionally set use_ema=False if value not specified (True never worked, and all configs except v1-inpainting-inference.yaml already correctly set it to False)
3 years ago
59c6511494
fix: fallback model_checkpoint if it's empty
...
This fixes the following error when SD attempts to start with a deleted checkpoint:
```
Traceback (most recent call last):
File "D:\Web\stable-diffusion-webui\launch.py", line 295, in <module>
start()
File "D:\Web\stable-diffusion-webui\launch.py", line 290, in start
webui.webui()
File "D:\Web\stable-diffusion-webui\webui.py", line 132, in webui
initialize()
File "D:\Web\stable-diffusion-webui\webui.py", line 62, in initialize
modules.sd_models.load_model()
File "D:\Web\stable-diffusion-webui\modules\sd_models.py", line 283, in load_model
checkpoint_info = checkpoint_info or select_checkpoint()
File "D:\Web\stable-diffusion-webui\modules\sd_models.py", line 117, in select_checkpoint
checkpoint_info = checkpoints_list.get(model_checkpoint, None)
TypeError: unhashable type: 'list'
```
3 years ago
bd81a09eac
fix support for 2.0 inpainting model while maintaining support for 1.5 inpainting model
3 years ago
ec5e072124
Merge pull request #4841 from R-N/vae-fix-none
...
Fix None option of VAE selector
3 years ago
1ed4f0e228
Depth2img model support
3 years ago
0376da180c
make it possible to save nai model using safetensors
3 years ago
dac9b6f15d
add safetensors support for model merging #4869
3 years ago
6074175faa
add safetensors to requirements
3 years ago
f108782e30
Merge pull request #4930 from Narsil/allow_to_load_safetensors_file
...
Supporting `*.safetensors` format.
3 years ago
1e506657e1
no-half support for SD 2.0
3 years ago
0efffbb407
Supporting `*.safetensors` format.
...
If a model file exists with extension `.safetensors` then we can load it
more safely than with PyTorch weights.
3 years ago
8662b5e57f
Merge branch 'a1111' into vae-fix-none
3 years ago
2c5ca706a7
Remove no longer necessary parts and add vae_file safeguard
3 years ago
c7be83bf02
Misc
...
Misc
3 years ago
abc1e79a5d
Fix base VAE caching was done after loading VAE, also add safeguard
3 years ago
eebf49592a
restore #4035 behavior
...
- if checkpoint cache is set to 1, keep 2 models in cache (current +1 more)
3 years ago
3b51d239ac
- do not use ckpt cache, if disabled
...
- cache model after is has been loaded from file
3 years ago
99043f3360
fix one of previous merges breaking the program
3 years ago
24fc05cf57
Merge branch 'master' into fix-ckpt-cache
3 years ago
3780ad3ad8
fix: loading models without vae from cache
3 years ago
fb3b564801
Merge branch 'master' into fix-ckpt-cache
3 years ago
f2a5cbe6f5
fix #3986 breaking --no-half-vae
3 years ago
056f06d373
Reload VAE without reloading sd checkpoint
3 years ago
f8c6468d42
Merge branch 'master' into vae-picker
3 years ago
brkirch
AUTOMATIC
Vladimir Mandic
Nicolas Patry
Yuval Aboulafia
linuxmobile ( リナックス )
MrCheeze
Dean van Dugteren
Jay Smith
Muhammad Rizqi Nur
cluder
digburn